science-of-finetuning / diffing-toolkitLinks
A toolkit that provides a range of model diffing techniques including a UI to visualize them interactively.
☆23Updated 2 weeks ago
Alternatives and similar repositories for diffing-toolkit
Users that are interested in diffing-toolkit are comparing it to the libraries listed below
Sorting:
- ☆18Updated 5 months ago
- ☆15Updated 3 months ago
- ☆81Updated 7 months ago
- Modified to support crosscoder training.☆23Updated 2 months ago
- ☆187Updated 2 months ago
- Unified access to Large Language Model modules using NNsight☆48Updated this week
- ☆348Updated last month
- ☆54Updated 10 months ago
- Mechanistic Interpretability Visualizations using React☆291Updated 9 months ago
- ☆242Updated last year
- ☆127Updated last year
- Tools for optimizing steering vectors in LLMs.☆11Updated 6 months ago
- ☆174Updated 10 months ago
- A library for efficient patching and automatic circuit discovery.☆77Updated 2 months ago
- ☆127Updated this week
- ☆106Updated 7 months ago
- ☆40Updated 2 weeks ago
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆42Updated last year
- Open source replication of Anthropic's Crosscoders for Model Diffing☆59Updated 11 months ago
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆215Updated last week
- ☆51Updated 2 months ago
- Sparse Autoencoder Training Library☆54Updated 5 months ago
- Attribution-based Parameter Decomposition☆31Updated 3 months ago
- ☆18Updated last month
- Steering vectors for transformer language models in Pytorch / Huggingface☆125Updated 7 months ago
- Evaluate interpretability methods on localizing and disentangling concepts in LLMs.☆56Updated last year
- Using sparse coding to find distributed representations used by neural networks.☆272Updated last year
- Sparse Autoencoder for Mechanistic Interpretability☆268Updated last year
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆221Updated 9 months ago
- Influence Functions with (Eigenvalue-corrected) Kronecker-Factored Approximate Curvature☆165Updated 3 months ago